Distinct accessory roles of Arabidopsis VEL proteins in Polycomb silencing.

Polycomb repressive complex 2 (PRC2) mediates epigenetic silencing of target genes in animals and plants. In Arabidopsis, PRC2 is required for the cold-induced epigenetic silencing of the FLC floral repressor locus to align flowering with spring. During this process, PRC2 relies on VEL accessory factors, including the constitutively expressed VRN5 and the cold-induced VIN3. The VEL proteins are physically associated with PRC2, but their individual functions remain unclear. Here, we show an intimate association between recombinant VRN5 and multiple components within a reconstituted PRC2, dependent on a compact conformation of VRN5 central domains. Key residues mediating this compact conformation are conserved among VRN5 orthologs across the plant kingdom. In contrast, VIN3 interacts with VAL1, a transcriptional repressor that binds directly to FLC These associations differentially affect their role in H3K27me deposition: Both proteins are required for H3K27me3, but only VRN5 is necessary for H3K27me2. Although originally defined as vernalization regulators, VIN3 and VRN5 coassociate with many targets in the Arabidopsis genome that are modified with H3K27me3. Our work therefore reveals the distinct accessory roles for VEL proteins in conferring cold-induced silencing on FLC, with broad relevance for PRC2 targets generally.

Structural basis of the interaction between BCL9-Pygo and LDB-SSBP complexes in assembling the Wnt enhanceosome.

The Wnt enhanceosome is responsible for transactivation of Wnt-responsive genes and a promising therapeutic target for treatment of numerous cancers with Adenomatous Polyposis Coli (APC) or ß-catenin mutations. How the Wnt enhanceosome is assembled remains poorly understood. Here we show that B-cell lymphoma 9 protein (BCL9), Pygopus (Pygo), LIM domain-binding protein 1 (LDB1) and single-stranded DNA-binding protein (SSBP) form a stable core complex within the Wnt enhanceosome. Their mutual interactions rely on a highly conserved N-terminal asparagine proline phenylalanine (NPF) motif of Pygo, through which the BCL9-Pygo complex binds to the LDB-SSBP core complex. Our crystal structure of a ternary complex comprising the N-terminus of human Pygo2, LDB1 and SSBP2 reveals a single LDB1-SSBP2 complex binding simultaneously to two Pygo2 molecules via their NPF motifs. These interactions critically depend on the NPF motifs which bind to a deep groove formed between LDB1 and SSBP2, potentially constituting a binding site for drugs blocking Wnt/ß-catenin signaling. Analysis of human cell lines lacking LDB or Pygo supports the functional relevance of the Pygo-LDB1-SSBP2 interaction for Wnt/ß-catenin-dependent transcription.

Head-to-tail polymerization by VEL proteins underpins cold-induced Polycomb silencing in flowering control.

Transcriptional silencing through the Polycomb silencing machinery utilizes a "read-write" mechanism involving histone tail modifications. However, nucleation of silencing and long-term stable transmission of the silenced state also requires P-olycomb Repressive Complex 2 (PRC2) accessory proteins, whose molecular role is poorly understood. The Arabidopsis VEL proteins are accessory proteins that interact with PRC2 to nucleate and propagate silencing at the FLOWERING LOCUS C (FLC) locus, enabling early flowering in spring. Here, we report that VEL proteins contain a domain related to an atypical four-helix bundle that engages in spontaneous concentration-dependent head-to-tail polymerization to assemble dynamic biomolecular condensates. Mutations blocking polymerization of this VEL domain prevent Polycomb silencing at FLC. Plant VEL proteins thus facilitate assembly of dynamic multivalent Polycomb complexes required for inheritance of the silenced state.

Plant vernalization proteins contain unusual PHD superdomains without histone H3 binding activity.

PHD fingers are modular domains in chromatin-associated proteins that decode the methylation status of histone H3 tails. A PHD finger signature is found in plant vernalization (VEL) proteins, which function as accessory factors of the Polycomb system to control flowering in Arabidopsis through an epigenetic silencing mechanism. It has been proposed that VEL PHD fingers bind to methylated histone H3 tails to facilitate association of the Polycomb silencing machinery with target genes. Here, we use structural analysis by X-ray crystallography to show that the VEL PHD finger forms the central module of a larger compact tripartite superdomain that also contains a zinc finger and a four-helix bundle. This PHD superdomain fold is only found in one other family, the OBERON proteins, which have multiple functions in Arabidopsis meristems to control plant growth. The putative histone-binding surface of OBERON proteins exhibits the characteristic three-pronged pocket of histone-binding PHD fingers and binds to methylated histone H3 tails. However, that of VEL PHD fingers lacks this architecture and exhibits unusually high positive surface charge. This VEL PHD superdomain neither binds to unmodified nor variously modified histone H3 tails, as demonstrated by isothermal calorimetry and NMR spectroscopy. Instead, the VEL PHD superdomain interacts with negatively charged polymers. Our evidence argues for evolution of a divergent function for the PHD superdomain in vernalization that does not involve histone tail decoding.

Selective function of the PDZ domain of Dishevelled in noncanonical Wnt signalling.

Dishevelled is a cytoplasmic hub that transduces Wnt signals to cytoplasmic effectors, which can be broadly characterised as canonical (ß-catenin dependent) and noncanonical, to specify cell fates and behaviours during development. To transduce canonical Wnt signals, Dishevelled binds to the intracellular face of Frizzled through its DEP domain and polymerises through its DIX domain to assemble dynamic signalosomes. Dishevelled also contains a PDZ domain, whose function remains controversial. Here, we use genome editing to delete the PDZ domain-encoding region from Drosophila dishevelled. Canonical Wingless signalling is entirely normal in these deletion mutants; however, they show defects in multiple contexts controlled by noncanonical Wnt signalling, such as planar polarity. We use nuclear magnetic resonance spectroscopy to identify bona fide PDZ-binding motifs at the C termini of different polarity proteins. Although deletions of these motifs proved aphenotypic in adults, we detected changes in the proximodistal distribution of the polarity protein Flamingo (also known as Starry night) in pupal wings that suggest a modulatory role of these motifs in polarity signalling. We also provide new genetic evidence that planar polarity relies on the DEP-dependent recruitment of Dishevelled to the plasma membrane by Frizzled.

Regulation of Dishevelled DEP domain swapping by conserved phosphorylation sites.

Wnt signals bind to Frizzled receptors to trigger canonical and noncanonical signaling responses that control cell fates during animal development and tissue homeostasis. All Wnt signals are relayed by the hub protein Dishevelled. During canonical (ß-catenin-dependent) signaling, Dishevelled assembles signalosomes via dynamic head-to-tail polymerization of its Dishevelled and Axin (DIX) domain, which are cross-linked by its Dishevelled, Egl-10, and Pleckstrin (DEP) domain through a conformational switch from monomer to domain-swapped dimer. The domain-swapped conformation of DEP masks the site through which Dishevelled binds to Frizzled, implying that DEP domain swapping results in the detachment of Dishevelled from Frizzled. This would be incompatible with noncanonical Wnt signaling, which relies on long-term association between Dishevelled and Frizzled. It is therefore likely that DEP domain swapping is differentially regulated during canonical and noncanonical Wnt signaling. Here, we use NMR spectroscopy and cell-based assays to uncover intermolecular contacts in the DEP dimer that are essential for its stability and for Dishevelled function in relaying canonical Wnt signals. These contacts are mediated by an intrinsically structured sequence spanning a conserved phosphorylation site upstream of the DEP domain that serves to clamp down the swapped N-terminal α-helix onto the structural core of a reciprocal DEP molecule in the domain-swapped configuration. Mutations of this phosphorylation site and its cognate surface on the reciprocal DEP core attenuate DEP-dependent dimerization of Dishevelled and its canonical signaling activity in cells without impeding its binding to Frizzled. We propose that phosphorylation of this crucial residue could be employed to switch off canonical Wnt signaling.

The deubiquitinase TRABID stabilizes the K29/K48-specific E3 ubiquitin ligase HECTD1.

Ubiquitin is a versatile posttranslational modification, which is covalently attached to protein targets either as a single moiety or as a ubiquitin chain. In contrast to K48 and K63-linked chains, which have been extensively studied, the regulation and function of most atypical ubiquitin chains are only starting to emerge. The deubiquitinase TRABID/ZRANB1 is tuned for the recognition and cleavage of K29 and K33-linked chains. Yet, substrates of TRABID and the cellular functions of these atypical ubiquitin signals remain unclear. We determined the interactome of two TRABID constructs rendered catalytic dead either through a point mutation in the catalytic cysteine residue or through removal of the OTU catalytic domain. We identified 50 proteins trapped by both constructs and which therefore represent candidate substrates of TRABID. The E3 ubiquitin ligase HECTD1 was then validated as a substrate of TRABID and used UbiCREST and Ub-AQUA proteomics to show that HECTD1 preferentially assembles K29- and K48-linked ubiquitin chains. Further in vitro autoubiquitination assays using ubiquitin mutants established that while HECTD1 can assemble short homotypic K29 and K48-linked chains, it requires branching at K29/K48 in order to achieve its full ubiquitin ligase activity. We next used transient knockdown and genetic knockout of TRABID in mammalian cells in order to determine the functional relationship between TRABID and HECTD1. This revealed that upon TRABID depletion, HECTD1 is readily degraded. Thus, this study identifies HECTD1 as a mammalian E3 ligase that assembles branched K29/K48 chains and also establishes TRABID-HECTD1 as a DUB/E3 pair regulating K29 linkages.

Feedback control of Wnt signaling based on ultrastable histidine cluster co-aggregation between Naked/NKD and Axin.

Feedback control is a universal feature of cell signaling pathways. Naked/NKD is a widely conserved feedback regulator of Wnt signaling which controls animal development and tissue homeostasis. Naked/NKD destabilizes Dishevelled, which assembles Wnt signalosomes to inhibit the ß-catenin destruction complex via recruitment of Axin. Here, we discover that the molecular mechanism underlying Naked/NKD function relies on its assembly into ultra-stable decameric core aggregates via its conserved C-terminal histidine cluster (HisC). HisC aggregation is facilitated by Dishevelled and depends on accumulation of Naked/NKD during prolonged Wnt stimulation. Naked/NKD HisC cores co-aggregate with a conserved histidine cluster within Axin, to destabilize it along with Dishevelled, possibly via the autophagy receptor p62, which binds to HisC aggregates. Consistent with this, attenuated Wnt responses are observed in CRISPR-engineered flies and human epithelial cells whose Naked/NKD HisC has been deleted. Thus, HisC aggregation by Naked/NKD provides context-dependent feedback control of prolonged Wnt responses.

Head-to-Tail Polymerization in the Assembly of Biomolecular Condensates.

Clustering of macromolecules is a fundamental cellular device underlying diverse biological processes that require high-avidity binding to effectors and substrates. Often, this involves a transition between diffuse and locally concentrated molecules akin to biophysical phase separation observable in vitro. One simple mechanistic paradigm underlying physiologically relevant phase transitions in cells is the reversible head-to-tail polymerization of hub proteins into filaments that are cross-linked by dimerization into dynamic three-dimensional molecular condensates. While many diverse folds and motifs can mediate dimerization, only two structurally distinct domains have been discovered so far to undergo head-to-tail polymerization, though these are widespread among all living kingdoms.

Limited dishevelled/Axin oligomerization determines efficiency of Wnt/ß-catenin signal transduction.

In Wnt/ß-catenin signaling, the transcriptional coactivator ß-catenin is regulated by its phosphorylation in a complex that includes the scaffold protein Axin and associated kinases. Wnt binding to its coreceptors activates the cytosolic effector Dishevelled (Dvl), leading to the recruitment of Axin and the inhibition of ß-catenin phosphorylation. This process requires interaction of homologous DIX domains present in Dvl and Axin, but is mechanistically undefined. We show that Dvl DIX forms antiparallel, double-stranded oligomers in vitro, and that Dvl in cells forms oligomers typically <10 molecules at endogenous expression levels. Axin DIX (DAX) forms small single-stranded oligomers, but its self-association is stronger than that of DIX. DAX caps the ends of DIX oligomers, such that a DIX oligomer has at most four DAX binding sites. The relative affinities and stoichiometry of the DIX-DAX interaction provide a mechanism for efficient inhibition of ß-catenin phosphorylation upon Axin recruitment to the Wnt receptor complex.

Stem cells can give rise to many types of specialized cells through a process called differentiation, which is partly regulated by changes in the levels of a protein known as ß-catenin. On one hand, a 'destruction complex' can keep ß-catenin levels low; this complex includes a protein called Axin and an enzyme known as GSK-3, which can tag ß-catenin for degradation. On the other hand, when ß-catenin levels need to increase, another protein called Dishevelled is activated. By binding to Axin, Dishevelled can bring the destruction complex in contact with other proteins, which leads to the deactivation of GSK-3. Dishevelled and Axin interact via a region that is similar in the two proteins, called DIX in Dishevelled and DAX in Axin. Studies of DIX and DAX have shown that both regions can form polymers ­ that is, a high number of similar units can bind together to form larger structures. However, these experiments were at higher concentrations than would be found in the cell. It was thought that, when combined, DIX and DAX might form these long chains together, preventing Axin from carrying out its role in destroying ß-catenin. Kan et al. set out to better understand this process by studying how DIX and DAX behave separately, and how they interact. The proteins were examined using a technique called cryo-electron microscopy, which allows scientists to dissect the structure of large proteins. When there was a high concentration of DIX in the sample, the molecules attached to one another to form long double-stranded helices. Similarly, DAX also formed helices, but these were shorter and only single-stranded. When the two proteins were combined, DAX bound only to the ends of short DIX chains, so that there are not more than four DAX chains attached to each DIX double helix. To see if this behaviour happens naturally, Kan et al. attached fluorescent tags to Dishevelled proteins and followed them in living cells: this showed that Dishevelled forms smaller chains with fewer than ten molecules. Together these results highlight how Dishevelled binds to Axin to deactivate GSK-3, to prevent the enzyme from promoting ß-catenin destruction. Mutations in the genes that encode ß-catenin or its regulators are associated with cancer. Ultimately, a better understanding of how ß-catenin is regulated could help to identify new opportunities for drug development.

Proteogenomics analysis unveils a TFG-RET gene fusion and druggable targets in papillary thyroid carcinomas.

Papillary thyroid cancer (PTC) is the most common type of endocrine malignancy. By RNA-seq analysis, we identify a RET rearrangement in the tumour material of a patient who does not harbour any known RAS or BRAF mutations. This new gene fusion involves exons 1-4 from the 5' end of the Trk fused Gene (TFG) fused to the 3' end of RET tyrosine kinase leading to a TFG-RET fusion which transforms immortalized human thyroid cells in a kinase-dependent manner. TFG-RET oligomerises in a PB1 domain-dependent manner and oligomerisation of TFG-RET is required for oncogenic transformation. Quantitative proteomic analysis reveals the upregulation of E3 Ubiquitin ligase HUWE1 and DUBs like USP9X and UBP7 in both tumor and metastatic lesions, which is further confirmed in additional patients. Expression of TFG-RET leads to the upregulation of HUWE1 and inhibition of HUWE1 significantly reduces RET-mediated oncogenesis.

DIX Domain Polymerization Drives Assembly of Plant Cell Polarity Complexes.

Cell polarity is fundamental for tissue morphogenesis in multicellular organisms. Plants and animals evolved multicellularity independently, and it is unknown whether their polarity systems are derived from a single-celled ancestor. Planar polarity in animals is conferred by Wnt signaling, an ancient signaling pathway transduced by Dishevelled, which assembles signalosomes by dynamic head-to-tail DIX domain polymerization. In contrast, polarity-determining pathways in plants are elusive. We recently discovered Arabidopsis SOSEKI proteins, which exhibit polar localization throughout development. Here, we identify SOSEKI as ancient polar proteins across land plants. Concentration-dependent polymerization via a bona fide DIX domain allows these to recruit ANGUSTIFOLIA to polar sites, similar to the polymerization-dependent recruitment of signaling effectors by Dishevelled. Cross-kingdom domain swaps reveal functional equivalence of animal and plant DIX domains. We trace DIX domains to unicellular eukaryotes and thus show that DIX-dependent polymerization is an ancient mechanism conserved between kingdoms and central to polarity proteins.

A direct heterotypic interaction between the DIX domains of Dishevelled and Axin mediates signaling to ß-catenin.

The Wnt-ß-catenin signaling pathway regulates embryonic development and tissue homeostasis throughout the animal kingdom. Signaling through this pathway crucially depends on the opposing activities of two cytoplasmic multiprotein complexes: the Axin destruction complex, which destabilizes the downstream effector ß-catenin, and the Dishevelled signalosome, which inactivates the Axin complex and thus enables ß-catenin to accumulate and operate a transcriptional switch in the nucleus. These complexes are assembled by dynamic head-to-tail polymerization of the DIX domains of Axin or Dishevelled, respectively, which increases their avidity for signaling effectors. Axin also binds to Dishevelled through its DIX domain. Here, we report the crystal structure of the heterodimeric complex between the two DIX domains of Axin and Dishevelled. This heterotypic interface resembles the interfaces observed in the individual homopolymers, albeit exhibiting a slight rearrangement of electrostatic interactions and hydrogen bonds, consistent with the heterotypic interaction being favored over the homotypic Axin DIX interaction. Last, cell-based signaling assays showed that heterologous polymerizing domains functionally substituted for the DIX domain of Dishevelled provided that these Dishevelled chimeras retained a DIX head or tail surface capable of binding to Axin. These findings indicate that the interaction between Dishevelled and Axin through their DIX domains is crucial for signaling to ß-catenin.

Rotational symmetry of the structured Chip/LDB-SSDP core module of the Wnt enhanceosome.

The Chip/LIM-domain binding protein (LDB)-single-stranded DNA-binding protein (SSDP) (ChiLS) complex controls numerous cell-fate decisions in animal cells, by mediating transcription of developmental control genes via remote enhancers. ChiLS is recruited to these enhancers by lineage-specific LIM-domain proteins that bind to its Chip/LDB subunit. ChiLS recently emerged as the core module of the Wnt enhanceosome, a multiprotein complex that primes developmental control genes for timely Wnt responses. ChiLS binds to NPFxD motifs within Pygopus (Pygo) and the Osa/ARID1A subunit of the BAF chromatin remodeling complex, which could synergize with LIM proteins in tethering ChiLS to enhancers. Chip/LDB and SSDP both contain N-terminal dimerization domains that constitute the bulk of their structured cores. Here, we report the crystal structures of these dimerization domains, in part aided by DARPin chaperones. We conducted systematic surface scanning by structure-designed mutations, followed by in vitro and in vivo binding assays, to determine conserved surface residues required for binding between Chip/LDB, SSDP, and Pygo-NPFxD. Based on this, and on the 4:2 (SSDP-Chip/LDB) stoichiometry of ChiLS, we derive a highly constrained structural model for this complex, which adopts a rotationally symmetrical SSDP2-LDB2-SSDP2 architecture. Integrity of ChiLS is essential for Pygo binding, and our mutational analysis places the NPFxD pockets on either side of the Chip/LDB dimer, each flanked by an SSDP dimer. The symmetry and multivalency of ChiLS underpin its function as an enhancer module integrating Wnt signals with lineage-specific factors to operate context-dependent transcriptional switches that are pivotal for normal development and cancer.

Bcl9 and Pygo synergise downstream of Apc to effect intestinal neoplasia in FAP mouse models.

Bcl9 and Pygo are Wnt enhanceosome components that effect ß-catenin-dependent transcription. Whether they mediate ß-catenin-dependent neoplasia is unclear. Here we assess their roles in intestinal tumourigenesis initiated by Apc loss-of-function (ApcMin), or by Apc1322T encoding a partially-functional Apc truncation commonly found in colorectal carcinomas. Intestinal deletion of Bcl9 extends disease-free survival in both models, and essentially cures Apc1322T mice of their neoplasia. Loss-of-Bcl9 synergises with loss-of-Pygo to shift gene expression within Apc-mutant adenomas from stem cell-like to differentiation along Notch-regulated secretory lineages. Bcl9 loss also promotes tumour retention in ApcMin mice, apparently via relocating nuclear ß-catenin to the cell surface, but this undesirable effect is not seen in Apc1322T mice whose Apc truncation retains partial function in regulating ß-catenin. Our results demonstrate a key role of the Wnt enhanceosome in ß-catenin-dependent intestinal tumourigenesis and reveal the potential of BCL9 as a therapeutic target during early stages of colorectal cancer.

Multiprotein complexes governing Wnt signal transduction.

Three multiprotein complexes have key roles in transducing Wnt signals from the plasma membrane to the cell nucleus - the ß-catenin destruction complex, or Axin degradasome, which targets the Wnt effector ß-catenin for proteasomal degradation in the absence of Wnt; the Wnt signalosome, assembled by polymerization of Dishevelled upon Wnt engaging its receptors, to inactivate the Axin degradasome, which allows ß-catenin to accumulate; and the Wnt enhanceosome which enables ß-catenin to gain access to target genes, to relieve their transcriptional repression by Groucho/TLE. This review focuses on recent advances that have highlighted mechanistic principles governing the assembly and function of these complexes.

Wnt-Dependent Inactivation of the Groucho/TLE Co-repressor by the HECT E3 Ubiquitin Ligase Hyd/UBR5.

Extracellular signals are transduced to the cell nucleus by effectors that bind to enhancer complexes to operate transcriptional switches. For example, the Wnt enhanceosome is a multiprotein complex associated with Wnt-responsive enhancers through T cell factors (TCF) and kept silent by Groucho/TLE co-repressors. Wnt-activated ß-catenin binds to TCF to overcome this repression, but how it achieves this is unknown. Here, we discover that this process depends on the HECT E3 ubiquitin ligase Hyd/UBR5, which is required for Wnt signal responses in Drosophila and human cell lines downstream of activated Armadillo/ß-catenin. We identify Groucho/TLE as a functionally relevant substrate, whose ubiquitylation by UBR5 is induced by Wnt signaling and conferred by ß-catenin. Inactivation of TLE by UBR5-dependent ubiquitylation also involves VCP/p97, an AAA ATPase regulating the folding of various cellular substrates including ubiquitylated chromatin proteins. Thus, Groucho/TLE ubiquitylation by Hyd/UBR5 is a key prerequisite that enables Armadillo/ß-catenin to activate transcription.

Constitutive scaffolding of multiple Wnt enhanceosome components by Legless/BCL9.

Wnt/ß-catenin signaling elicits context-dependent transcription switches that determine normal development and oncogenesis. These are mediated by the Wnt enhanceosome, a multiprotein complex binding to the Pygo chromatin reader and acting through TCF/LEF-responsive enhancers. Pygo renders this complex Wnt-responsive, by capturing ß-catenin via the Legless/BCL9 adaptor. We used CRISPR/Cas9 genome engineering of Drosophila legless (lgs) and human BCL9 and B9L to show that the C-terminus downstream of their adaptor elements is crucial for Wnt responses. BioID proximity labeling revealed that BCL9 and B9L, like PYGO2, are constitutive components of the Wnt enhanceosome. Wnt-dependent docking of ß-catenin to the enhanceosome apparently causes a rearrangement that apposes the BCL9/B9L C-terminus to TCF. This C-terminus binds to the Groucho/TLE co-repressor, and also to the Chip/LDB1-SSDP enhanceosome core complex via an evolutionary conserved element. An unexpected link between BCL9/B9L, PYGO2 and nuclear co-receptor complexes suggests that these ß-catenin co-factors may coordinate Wnt and nuclear hormone responses.